Statistical evaluation of biogeochemical variables affecting spatiotemporal distributions of multiple free metal ion concentrations in an urban estuary

Free metal ion concentrations have been recognized as a better indicator of metal bioavailability in aquatic environments than total dissolved metal concentrations. However, our understanding of the determinants of free ion concentrations, especially in a metal mixture, is limited, due to underexplored techniques for measuring multiple free metal ions simultaneously. In this work, we performed statistical analyses on a large dataset containing repeated measurements of free ion concentrations of Cu, Zn, Pb, Ni, and Cd, the most commonly measured metals in seawater, at five inshore locations in Boston Harbor, previously collected using an in-situ equilibrium-based multi-metal free ion sampler, the 'Gellyfish'. We examined correlations among these five metals by season, and evaluated effects of 10 biogeochemical variables on free ion concentrations over time and location through multivariate regressions. We also explored potential clustering among the five metals through a principal component analysis. We found significant correlations among metals, with varying patterns over season. Our regression results suggest that instead of dissolved metals, pH, salinity, temperature and rainfall were the most significant determinants of free metal ion concentrations. For example, a one-unit decrease in pH was associated with a 2.2 (Cd) to 99 (Cu) times increase in free ion concentrations. This work is among the first to reveal key contributors to spatiotemporal variations in free ion concentrations, and demonstrated the usefulness of the Gellyfish sampler in routine sampling of free ions within metal mixtures and in generating data for statistical analyses.